Background: Low birth weight (LBW) has been associated with an increased risk of cardiovascular disease (CVD). A previous study, however, found higher risk of atrial fibrillation (AF) in individuals with higher birth weight (BW). To further understand this apparent paradox, we examined the relationship between AF and BW in the Atherosclerosis Risk in Communities (ARIC) cohort. Methods: The analysis included 10,132 individuals free of AF at baseline (1996–1998), who provided BW information, were not born premature, and were not a twin. Self-reported BW was categorized as low (<2.5 kg), medium (2.5-4 kg), and high (>4.0 kg). AF incidence was ascertained from hospital discharge codes and death certificates. We used multivariable Cox proportional hazard models to determine the hazard ratios (HR) and 95% confidence intervals (CI) of AF across BW groups. Results: During an average follow-up of 10.3 years, we identified 882 incident AF cases. LBW was associated with higher risk of AF. Compared to individuals in the medium BW category, the HR (95% CI) of AF was 1.33 (0.99, 1.78) for LBW and 1.00 (0.81, 1.24) for high BW after adjusting for sociodemographic variables (p for trend = 0.29). Additional adjustment for CVD risk factors did not attenuate the associations (HR 1.42, 95% CI 1.06, 1.90 for LBW and HR 0.86, 95% CI 0.69-1.07 for high BW, compared to medium BW, p for trend = 0.01). Conclusion: LBW was associated with a higher risk of AF. This association was independent of known predictors of AF and is consistent with that observed for other cardiovascular diseases.